Purpose:

The purpose of this study is to evaluate the findings of the radiography, computed tomography and magnetic resonance imaging in patients with bacterial spondylitis and to reveal superiority of these imaging moda-lities to each other.

Materials and methods:

11 cases with tuberculous spondylitis, 24 cases with brucella spondylitis and 5 cases with postoperative spondylitis were examined with the radi-ography, CT and MRI. The location and number of the in-volved vertebrae, end-plate irregularity, bone destruction and sclerosis, decrease in height of vertebral body, spinal deformity (scoliosis, kyphosis), narrowing or presence of gas in the intervertebral disc space and paravertebral soft tissue changes were evaluated. In addition, spinal cord, thecal sac and nerve root compressions were evaluated in CT and MRI images.

Result:

According to MRI findings, the most involved vertebra in tuberculous, brucellar and postoperative spondylitis was L4 (%23), L5 (%20) and L5 (%36) respec-tively. Lumbar region was the most affected site througho-ut vertebral column in all these three groups. The total number of the involved vertebrae was 97 in MRI, 88 in CT, 68 in the radiography. In evaluating paravertebral soft tissue involvement MRI was found to be superior to radiography (Kappa: 0,08 P<0,001) and CT (Kappa: 0,39 p<0,01), in evaluating epidural involvement MRI was found to be supe-rior to CT (Kappa: 0,38 p<0,01). There was no difference between MRI and CT in evaluating thecal sac, spinal cord and nerve root compression (p>0,05).

Conclusion:

The role of the radiography in the diagno-sis of spinal infections is very limited since it is normal in early stages of the disease; pathologic findings appear in late stages and are nonspesific. Paravertebral soft tissue changes and epidural involvement are best evaluated with MRI. Compression of the spinal cord, thecal sac and nerve root can be evaluated either with CT or MRI.

Keywords: Spine, infection, radiography, CT, MRI.